Use of ultra-sensitive whole genome circulating tumor DNA detection to predict early recurrence in high-risk localized RCC after nephrectomy: Preliminary real-world findings.

Abstract

516 Background: Circulating tumor DNA (ctDNA) has emerged as a powerful tool for detecting molecular residual disease (MRD) across multiple solid tumors, but its application in renal cell carcinoma (RCC) has historically been limited by low ctDNA shedding. Recent advances in ultra-sensitive, tumor-informed whole-genome sequencing (WGS) assays capable of tracking up to 1,800 patient-specific variants now enable ctDNA detection at concentrations as low as ~ 1 part per million (PPM). This study evaluates the prognostic value of ultra-sensitive ctDNA following nephrectomy in high-risk localized RCC. Methods: This single-center retrospective cohort study included patients with localized RCC who underwent at least one ctDNA assessment using the Next Personal Dx WGS tumor-informed assay. ctDNA testing was performed within 3 months after nephrectomy (3-month landmark), and/or at 6 months (≥3 to < 7 months) post-surgery. The primary endpoint was recurrence rate based on ctDNA status at any timepoint. A secondary endpoint was recurrence rate in patients with persistent ctDNA negativity, defined as undetectable ctDNA at both landmarks assessments. Results: Twenty-two patients (42 plasma samples) were included, with a median follow-up of 7.0 months (range, 3.2 – 19.3). The cohort was predominantly male (77%), with a median age of 63 years and largely white. A total of 87% underwent radical nephrectomy, 95% had pT3 disease, with a median size of 6.4 cm, all patients were pN0/Nx, 91% clear cell, 9% unclassified, 18% G2, 50% G3 and, 32% G4, 14% had sarcomatoid features and 50% received adjuvant therapy. Median time to the 3-month landmark ctDNA draw and subsequent 6-month assessment was 1.4 and 4.8 months, respectively. At the landmark timepoint, 95% of patients had ctDNA collected, and 91% had 6-month testing. Three patients (14%) had ctDNA positivity at any timepoint (two persistent and one conversion, all < 100ppm). Two of these developed radiographic recurrence (1 and 4 months after first ctDNA detection). In contrast, only 1 of 20 patients with 3-month landmark ctDNA negativity recurred, and none (0/16) of those with persistent ctDNA negativity recurred during the follow-up. Conclusions: Ultra-sensitive ctDNA appears promising for identifying RCC patients at higher risk for recurrence, while persistent undetectable ctDNA correlates with a low observed risk. Prospective studies are warranted to test its use in disease monitoring or patient selection for adjuvant treatment.